Github user nongli commented on a diff in the pull request:
https://github.com/apache/spark/pull/10974#discussion_r51294294
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/misc.scala
---
@@ -365,91 +391,48 @@ case class Murmur3Hash(children: Seq[Expression],
seed: Int) extends Expression
hashLong(s"$input.toUnscaledLong()")
} else {
val bytes = ctx.freshName("bytes")
- val code = s"byte[] $bytes =
$input.toJavaBigDecimal().unscaledValue().toByteArray();"
- val offset = "Platform.BYTE_ARRAY_OFFSET"
- val result = s"$hasher.hashUnsafeBytes($bytes, $offset,
$bytes.length, $seed)"
- ExprCode(code, "false", result)
+ s"""
+ final byte[] $bytes =
$input.toJavaBigDecimal().unscaledValue().toByteArray();
+ ${hashBytes(bytes)}
+ """
}
case CalendarIntervalType =>
- val microsecondsHash = s"$hasher.hashLong($input.microseconds,
$seed)"
- val monthsHash = s"$hasher.hashInt($input.months,
$microsecondsHash)"
- inlineValue(monthsHash)
- case BinaryType =>
- val offset = "Platform.BYTE_ARRAY_OFFSET"
- inlineValue(s"$hasher.hashUnsafeBytes($input, $offset,
$input.length, $seed)")
+ val microsecondsHash = s"$hasher.hashLong($input.microseconds,
$result)"
+ s"$result = $hasher.hashInt($input.months, $microsecondsHash);"
+ case BinaryType => hashBytes(input)
case StringType =>
val baseObject = s"$input.getBaseObject()"
val baseOffset = s"$input.getBaseOffset()"
val numBytes = s"$input.numBytes()"
- inlineValue(s"$hasher.hashUnsafeBytes($baseObject, $baseOffset,
$numBytes, $seed)")
+ s"$result = $hasher.hashUnsafeBytes($baseObject, $baseOffset,
$numBytes, $result);"
- case ArrayType(et, _) =>
- val result = ctx.freshName("result")
+ case ArrayType(et, containsNull) =>
val index = ctx.freshName("index")
- val element = ctx.freshName("element")
- val elementHash = computeHash(element, et, result, ctx)
- val code =
- s"""
- int $result = $seed;
- for (int $index = 0; $index < $input.numElements(); $index++) {
- if (!$input.isNullAt($index)) {
- final ${ctx.javaType(et)} $element = ${ctx.getValue(input,
et, index)};
- ${elementHash.code}
- $result = ${elementHash.value};
- }
- }
- """
- ExprCode(code, "false", result)
+ s"""
+ for (int $index = 0; $index < $input.numElements(); $index++) {
+ ${nullSafeElementHash(input, index, containsNull, et, result,
ctx)}
+ }
+ """
- case MapType(kt, vt, _) =>
- val result = ctx.freshName("result")
+ case MapType(kt, vt, valueContainsNull) =>
val index = ctx.freshName("index")
val keys = ctx.freshName("keys")
val values = ctx.freshName("values")
- val key = ctx.freshName("key")
- val value = ctx.freshName("value")
- val keyHash = computeHash(key, kt, result, ctx)
- val valueHash = computeHash(value, vt, result, ctx)
- val code =
- s"""
- int $result = $seed;
- final ArrayData $keys = $input.keyArray();
- final ArrayData $values = $input.valueArray();
- for (int $index = 0; $index < $input.numElements(); $index++) {
- final ${ctx.javaType(kt)} $key = ${ctx.getValue(keys, kt,
index)};
- ${keyHash.code}
- $result = ${keyHash.value};
- if (!$values.isNullAt($index)) {
- final ${ctx.javaType(vt)} $value = ${ctx.getValue(values,
vt, index)};
- ${valueHash.code}
- $result = ${valueHash.value};
- }
- }
- """
- ExprCode(code, "false", result)
+ s"""
+ final ArrayData $keys = $input.keyArray();
+ final ArrayData $values = $input.valueArray();
+ for (int $index = 0; $index < $input.numElements(); $index++) {
+ ${nullSafeElementHash(keys, index, false, kt, result, ctx)}
+ ${nullSafeElementHash(values, index, valueContainsNull, vt,
result, ctx)}
+ }
+ """
case StructType(fields) =>
- val result = ctx.freshName("result")
- val fieldsHash = fields.map(_.dataType).zipWithIndex.map {
- case (dt, index) =>
- val field = ctx.freshName("field")
- val fieldHash = computeHash(field, dt, result, ctx)
- s"""
- if (!$input.isNullAt($index)) {
- final ${ctx.javaType(dt)} $field = ${ctx.getValue(input,
dt, index.toString)};
- ${fieldHash.code}
- $result = ${fieldHash.value};
- }
- """
+ fields.zipWithIndex.map { case (field, index) =>
+ nullSafeElementHash(input, index.toString, field.nullable,
field.dataType, result, ctx)
}.mkString("\n")
- val code =
- s"""
- int $result = $seed;
- $fieldsHash
- """
- ExprCode(code, "false", result)
- case udt: UserDefinedType[_] => computeHash(input, udt.sqlType,
seed, ctx)
+ case udt: UserDefinedType[_] => computeHash(input, udt.sqlType,
result, ctx)
--- End diff --
This is a bug fix right? Do we have tests for this?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
